Large vocabulary continuous speech recognition with context-dependent DBN-HMMS

George E. Dahl, Dong Yu, Li Deng, Alex Acero

Abstract

The context-independent deep belief network (DBN) hidden Markov model (HMM) hybrid architecture has recently achieved promising results for phone recognition. In this work, we propose a context-dependent DBN-HMM system that dramatically outperforms strong Gaussian mixture model (GMM)-HMM baselines on a challenging, large vocabulary, spontaneous speech recognition dataset from the Bing mobile voice search task. Our system achieves absolute sentence accuracy improvements of 5.8% and 9.2% over GMM-HMMs trained using the minimum phone error rate (MPE) and maximum likelihood (ML) criteria, respectively, which translate to relative error reductions of 16.0% and 23.2%.
BibTeX
@inproceedings{dahl2011large,
  author = {George E. Dahl and Dong Yu and Li Deng and Alex Acero},
  title = {{Large vocabulary continuous speech recognition with context-dependent DBN-HMMS}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2011},
  doi = {10.1109/icassp.2011.5947401},
}

Loading related content...

Figure